BriefGPT.xyz
Apr, 2024
不完全统计扰动：私有最小二乘的稳定估计器
Insufficient Statistics Perturbation: Stable Estimators for Private Least Squares
HTML
PDF
Gavin Brown, Jonathan Hayase, Samuel Hopkins, Weihao Kong, Xiyang Liu...
TL;DR
我们提出了一个针对普通最小二乘问题的样本和时间高效的差分隐私算法，误差线性依赖于维度并且与$X^ op X$的条件数无关，其中$X$是设计矩阵。我们的算法具有接近最优的准确性保证，适用于具有有限统计杠杆率和有界残差的任何数据集。
Abstract
We present a sample- and time-efficient
differentially private algorithm
for
ordinary least squares
, with error that depends linearly on the
dime
→